I know far too many developers who cringe at the thought of continuous deployment. Thoughts of broken code and an endless stream of customer support requests puts continuous deployment into the “nice to have, but not right now” pile.
But, it might just be a case of clearing the fog on process so you get visibility into the deployment pipeline. If your team can get more visibility into releases, you will have more confidence your shipped code is quality. With more visibility, you’ll be deploying faster and earlier before you know it.
Shipping a big release is taxing on your QA and testing teams and the pressure to get every part of the build correct is huge. Continuous deployment allows you to deploy smaller chunks of code more frequently so you have more control over the release.
Think “shipping responsibly.”
With that in mind, it’s easy to see why a lack of visibility can hinder continuous deployment’s goal of shipping more code to a tight schedule.
Best practices for your software development lifecycle (SDLC) demand that you know what’s going on in your code. And if you are responsible for your release process, you are probably held accountable for what customers receive.
Production demands visibility into code quality; otherwise, you risk economic damage to your business in the form of bugs and crashes. With this in mind, understandably, many teams hesitate to continuously deploy thorough lack of confidence in the quality of shippable code through the build process. If more visibility means more confidence, let’s take a look at the stages in a build process and how to increase visibility at each stage.
Continuous deployment will require resources and modern tooling initiated by your best people. Although budget often can be one of the biggest roadblocks to continuous deployment’s success, don’t withhold funds. Invest in the tools and people first, and the process will come. Modern tooling that is updated regularly will give you the best of what’s available, allowing you to effectively see what’s happening in each stage from source control to deployment and beyond.
You may not want your continuous deployment to be fully automatic. There are risks associated with continuous deployment, and to mitigate those risks you need a human to push the “deploy” button. It’s more about making your team feel accountable for their actions. The developer sees the change go live and is the first to test them.
Do you have key stakeholders waiting for updates on feature releases? Rush boxes might be the answer. A rush box is a server used for work that is not ready for the QA process. This will give anyone to give feedback and authorize changes.
Continuous deployment is certainly not without its risks. But, if you incorporate a culture of visibility into your processes, you can quickly identify where the gaps are so you can start to scale. Your toolkit will generate a huge amount of transactional data that will help to steer your company in the right direction.
It amazes me how many companies still push out changes and rely on the end user to tell them whether the deployment was successful. Defects in production code have economic impacts on your company, and undiscovered bugs and errors can rack up into technical debt. But, you can’t change what you can’t see. Small changes are always going into production, so time for QA and regression tests naturally lessens. Therefore, you need to know what happened, and when, which is where a monitoring tool such as crash reporting can help. I’ve written before about the benefits of monitoring in continuous deployment, and it’s an integral step that is often missing. Monitoring can also help you make informed decisions about when to rollback a deployment.
Rollbacks have an economic impact to your business, much in the way errors do. However, if a decision to roll back is data-driven, it’s an easy call for a senior developer to make.
Don’t let the lack of visibility into your deployment pipeline affect your decision to implement a continuous deployment process. Invest in people and tooling first, and the process and scale will come.
Cognition Labs' Devin is creating a lot of buzz in the industry, but John Willis urges organizations to proceed with…
While most app developers work for organizations that have platform teams, there isn't much consistency regarding where that team reports.
Day Two DevOps is a phase in the SDLC that focuses on enhancing, optimizing and continuously improving the software development…
A global survey of 500 IT professionals suggests organizations are not making a lot of progress in their ability to…
In part five of this series, hosts Alan Shimel and Mitch Ashley are joined by Bryan Cole (Tricentis), Ixchel Ruiz…
There has never been a better time to be a software developer. There is a language and framework to solve…